Re: [MMUSIC] BUNDLE DECISION Q6: Do we always mandate 2 Offer/Answers during session establishment?

[Paul Kyzivat <[e-mail]>]

On 11/13/13 2:43 PM, Suhas Nandakumar wrote:
> Hi Christer
>
>    If it helps, I might point you to the Taxonomy document to see if
> there is a terminology from that document that better fits the purpose
> for naming here
>
> http://tools.ietf.org/html/draft-ietf-avtext-rtp-grouping-taxonomy-00

I don't think the taxonomy is particularly helpful here.
(Maybe it should be.)

The concept we are after is at the extreme upper boundary of what the 
taxonomy considers. It has a definition of multimedia session, and also 
mentions definitions from 4566 and 3550. They are all quite similar, but 
not identical.

Our use here is, I think, roughly a subset of the 4566 definition, where 
there are only two participants. But it might not actually be so. It is 
unclear in those definitions whether the participants are the senders 
and receivers of signaling, or the senders and receivers of media. 
Often, but not always, they are the same. For the purposes of this draft 
we want to be talking about the senders and receivers of the SDP offers 
and answers.

It may be necessary to create a definition for the purpose.

	Thanks,
	Paul

> Cheers
> Suhas
>
>
> On Wed, Nov 13, 2013 at 11:32 AM, Christer Holmberg
> <[e-mail] <mailto:[e-mail]>>
> wrote:
>
>     Hi,
>
>      >>> This works for me except for one thing: the word "session",
>     which I fear is sufficiently ambiguous to present trouble.
>      >>>
>      >>> I suggest using "multi-media session". (We can't use "SIP
>     session" or
>      >>> "SIP dialog" because this might be used without SIP.)
>      >>
>      >> I agree regarding not being able to use SIP terminology.
>      >>
>      >> I guess "multi-media session" is one option. What about "SDP
>     session"?
>      >
>      > I thought about that. I'm not in love with multi-media session.
>      >
>      > When I look at 4566 I find a definition of "session" as
>     multi-media session. I think "SDP session" gets the point across,
>     though I have not seen "SDP session" used anywhere else.
>
>     We could even say "SDP session (e.g. a SIP dialog)"
>
>     Regards,
>
>     Christer
>
>
>      > On 11/12/13 3:39 PM, Christer Holmberg wrote:
>      >> Hi,
>      >>
>      >> In Vancouver we decided that an Offerer MUST NOT assign a shared
>     address (with a non-zero port) to multiple m- lines until the
>     Answerer, within the session, has indicated support of BUNDLE.
>      >>
>      >> I suggest the following piece of text to implement  the decision
>     in the BUNDLE spec:
>      >>
>      >>      "The Offerer MUST NOT assign a shared address with a
>     non-zero port
>      >>              value to multiple "m=" lines until it has, within
>     the given session,
>      >>              received an SDP Answer indicating that the Answerer
>     supports the
>      >>              BUNDLE mechanism."
>      >>
>      >> Note that there will be specific text regarding the usage of
>     port zero for 'bundle-only' m- lines.
>      >>
>      >> Regards,
>      >>
>      >> Christer
>      >> _______________________________________________
>      >> mmusic mailing list
>      >> [e-mail] <mailto:[e-mail]>
>      >> https://www.ietf.org/mailman/listinfo/mmusic
>      >>
>      >
>      > _______________________________________________
>      > mmusic mailing list
>      > [e-mail] <mailto:[e-mail]>
>      > https://www.ietf.org/mailman/listinfo/mmusic
>      >
>
>     _______________________________________________
>     mmusic mailing list
>     [e-mail] <mailto:[e-mail]>
>     https://www.ietf.org/mailman/listinfo/mmusic
>
>